New Technique Promises Much Faster Hard Drive Write Speeds 148
MrSeb writes "Hold onto your hats: Scientists at the University of York, England have completely rewritten the rules of magnetic storage (abstract; full paper paywalled). Instead of switching a magnetic region using a magnetic field (like a hard drive head), the researchers have managed to switch a ferrimagnetic nanoisland using a 60-femtosecond laser. Storing magnetic data using lasers is up to 1,000 times faster than writing to a conventional hard drive (we're talking about gigabytes or terabytes per second) — and the ferrimagnetic nanoislands that store the data are capable of storage densities that are some 15 times greater than existing hard drive platters. Unfortunately the York scientists only detailed writing data with lasers; there's no word on how to read it."
write only media FTW! (Score:5, Funny)
Who needs to read data back anyway?
Re:write only media FTW! (Score:5, Funny)
Re:write only media FTW! (Score:5, Funny)
Re:write only media FTW! (Score:5, Funny)
Re:write only media FTW! (Score:5, Funny)
Cryogenic data storage (Score:4, Funny)
This is the data equivalent of freezing Walt Disney and assuming that someday we'll figure out how to thaw and revive him. Write now, read someday.
Re:Cryogenic data storage (Score:4, Funny)
Re:Cryogenic data storage (Score:5, Insightful)
Re: (Score:2)
No, the answer would be "No" because the question should have been about "Steamboat Willie".
Re:Cryogenic data storage (Score:5, Funny)
Re: (Score:2)
Oh jeez why don't I get mod points anymore? XD
Re: (Score:2)
It's a Family Guy quote, calm down.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
In large disk subsystems used for archival, the assumption is generally that >90% of the data is never read back. That's why they're heavily write-optimized. /dev/null.
Now if we could determine which data isn't needed any more, we could store it on
Re: (Score:2)
In large disk subsystems used for archival, the assumption is generally that >90% of the data is never read back. That's why they're heavily write-optimized. /dev/null.
Now if we could determine which data isn't needed any more, we could store it on
I once worked at a place that had a hierarchical storage system that did pretty much just that -- little used data was migrated off of expensive disks to tape. However, due to a a flaw in the tape drive or the tapes themselves, over time the data became unreadable, leaving much of the data that was migrated off of disk unreadable. Most of it was recovered from backups (which fortunately used a completely different tape system).
Do companies still use HSM that moves data to tape or is everyone using cheap SAT
Re: (Score:2)
Both. I used to do tape but I'm involved with disk based backup now, with SSDs, 15k SAS drives and 7200rpm SATA. But disk based backup systems get full so you still have to offload old data to tape, hopefully to never touch it again. That means tape better be reliable, it's the last resort. Customer escalations because they can't read their data are no fun. The good thing is that low end tape libraries are dying out - the $5k ones that were bought by customers with clueless IT organizations. Companies that
So how do they know if they actually wrote it (Score:2, Interesting)
If they can't read it, how do they know if they actually wrote it? Or maybe reading it is 10,000 times slower than current read technology.
Re:So how do they know if they actually wrote it (Score:5, Informative)
It's stored in the same way as a normal hard disk - in ferromagnetic domains on a platter. You can still read it back using the same techniques as current drives (i.e. put a coil over it and see which way the induced current flows), but you then have a drive that you can write to orders of magnitude than you can read from it. I can think of a few places where this might be useful. The most obvious is the underlying storage for something like ZFS. For reliability, you want to flush everything to the backing store as quickly as possible, and with copy-on-write and snapshotting you may never erase it, but most of your reads are satisfied from flash or DRAM caches. A drive using this technology would let you dump data there as quickly as you wanted and would let you read it back for data recovery if you needed to, while in normal operation you wouldn't care about the read speed because reading from the disk is comparatively rare. It would also be useful for a number of scientific applications. I did some work a few years ago with someone building a solar observatory. A single one of their cameras generated 10GB/s of data, and they had 8 cameras in a typical setup. They run these for the entire time that the sun is visible. A single drive that can handle a sustained write speed of 1GB/s would be very useful for them (although they'd fill up several per hour...).
For consumer devices, random read speed is still the most important factor, and mechanical drives suck at that.
Re: (Score:3, Insightful)
Re: (Score:2)
I don't think that when storing such absurd amounts of data as high speed, you're interested in possible fragmentation.
Re:So how do they know if they actually wrote it (Score:5, Interesting)
It's time to dust off the old concept of hard sectored discs ;) Realistically, of course, it's a bit more complex than that.
First of all, modern hard drives have a servo track that's used to maintain radial position of the head servo. Instead of each hard drive having a very accurate (and expensive) radial and axial head position sensor, you pay for it once, install it in the factory, use it to accurately guide a hard drive to write the servo track. Its cost is amortized over thousands of drives made. This is probably the reason for a covered up radial slot in many hard drive enclosures: I guess it's used for the sensor to couple with the head system while the drive writes the servo track. Or perhaps the servo platter is prewritten outside the disc? Someone familiar with how it's made please chip in!
The servo track can be also used to provide angular position feedback. A rough estimate of angular position of the spindle is available first from the Hall sensors in the spindle motor. A somewhat more accurate estimate can be had from back-EMF from the spindle motor windings. This still is methinks a couple orders of magnitude away from what's needed to pack sectors tightly on the drive -- thus the feedback can come from the servo track. Not having to read the data tracks helps with packing the sectors: there's no read-write switchover overhead (if it were significant -- perhaps it isn't nowadays). The servo head is always reading, and the data heads can be kept in write/erase standby. It'd be nondestructive, but read amplifiers are disconnected to prevent saturating them -- amplifier overload recovery is slow. Heck, if you want an amp that recovers from overloads quickly, you have to split it into more stages, and you need fast clamps between each stage. There are other similar approaches to this problem, too, and perhaps modern read amps are designed to deal with overloads gracefully -- I never tested a recent one. Stuff from a decade ago was painfully slow on overloads (tried to reuse a head amp from a hard drive for a non-drive-related project).
Alas, this ultra-fast-writing drive would unfortunately need very accurate position sensors -- both angular and radial. It's an engineering issue to make those affordable, as is the design of the optochip with femtosecond laser and its driver and serializer. The latter would probably take a couple serial lanes and multiplex them -- I presume it's not all that easy to push 10gbit/s data between external chips and the laser driver/laser combo. I think that to make it all practical you need an on-chip serializer, write precompensation, driver, and the diode. Perhaps the diode would be "tacked on" later to a substrate that has everything else. I only imagine that bond wire parasitics, even over a couple mm, become kinda important when the laser waveform has a 100GHz bandwidth...
Re: (Score:2)
It's not too abstruse to see that solid-state devices are the natural way forward.
Re:So how do they know if they actually wrote it (Score:4, Insightful)
Why? If they can write TB/s and store data at 15X of current capacity, and SSDs can't, why move to SSDs?
The read problem is easily resolved by having multiple read heads that can read independently.
Re: (Score:2)
Why? If they can write TB/s
Except that youre still limited by rotational latency and whatnot. Was the magnetic write head ever the main bottleneck?
Re: (Score:3)
Was the magnetic write head ever the main bottleneck?
Maybe not the 'main' bottleneck, but it depends on the application, no? Seems to me there are at least a few firehose situations where you can never have enough write bandwidth (say, uncompressed video-capture).
Maybe normal workloads on normal filesystems wouldn't see much improvement, but I bet you could find ways to capitalise on the extra bandwidth and space. Log-structured filesystems spring to mind for one.
Re: (Score:2)
Centralized backup, especially of large data-stores. You have to write massive amounts of data on a regular basis, but rarely read the data, and when you do you usually only need a small subset of what's been written. I could imagine it being useful for certain kinds of RAID configurations and networ
Re: (Score:2)
I don't care which way becomes the "best" choice as long as both styles interface through a standardized connector.
Both sides of chip vs platter will always have their own strengths and weaknesses, I like choice.
It's very easy to see this becoming the highest cost and highest performance drive of the near future that server admins and performance enthusiasts go to. While the SSD takes over as the PC and small device storage of choice.
Re: (Score:2)
Re: (Score:2)
Re:So how do they know if they actually wrote it (Score:5, Informative)
Re: (Score:1)
Readback can be easier (Score:5, Interesting)
This solves a major problem with mag recording. Readback head have always been way smaller than write head. You can read back with just a tiny permalloy head but to write you need large currents and loops of wire. So miniaturization has been limited by the write head size not the read head. This solves the write-head size problem but may have created a new read head problem. But that's very promising.
No you can't (Score:3)
This uses ferrimagnetic domains, not ferromagnetic domains. There is no external magnetic field, and you can't use a coil to read them.
Re: (Score:2)
Re: (Score:2)
As I understand, if you can accurately write such a smaller magnetic domain with a laser vs. the relatively large area under a write head, you obviously increase the data density. And since higher and more focused energy to flip the domains can now be applied, lessening the problem of flipping their neighbors, this probably means smaller particle and higher coercivity media can be used or developed. This also implies that the tracks get smaller and reduces or eliminates guard areas and tracks. All of which
Re: (Score:2)
You don't know if you actually wrote successfully on today's disk drives either.
At last! (Score:5, Funny)
omg (Score:4, Funny)
frickin hard drives with laser beams!
Re:omg (Score:5, Funny)
1) Western Digital licenses this technology.
2) WD releases a new model, the WD Shark. A "Now with LASERs!" starburst graphic adorns the front of the box.
3) Profit!
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
They could make a thumb drive using a miniature hard drive and call it the Sharktooth.
...okay WD, cut me a check already, would you?
Re: (Score:2)
2a) And a warning sticker on the back that says "Do not look into hard drive with remaining eye."
Re: (Score:2)
You mean like, I don't know.....cd roms as hard drives? or DVD's? or Laserdisc?
Re: (Score:2)
No, the data is not stored in a magnetic domain on an optical disk.
Re: (Score:2)
why go so far? magneto opticals work too
Memory Hole (Score:2)
Re:Memory Hole (Score:4, Funny)
Goatse Drive technology?
Yay! if they figure out 2 read da stuff, we'r gold (Score:1)
writing the cute useless powerpoint presentations that waste so much everyone's time will be done 1000 times faster, so will downloading swimsuit pictures (minus the swimsuit for some :) )
awesome, we're gonna be able to waste time so much faster haha
Re: (Score:1)
Re: (Score:1)
Implied Read? (Score:1)
How can scientists know the write was successful without being able to read back as well...surely there is an in implied read in the mix, otherwise the discovery isn't worth the paper it is written on!
Re: (Score:2)
Who cares about reading? (Score:5, Insightful)
Re: (Score:3, Insightful)
You can only write as fast as data can be read so your backups will not be 1,000 times faster.
Re: (Score:2)
You say that until you've got to be at the office until 3am waiting for a backup to restore for DR.
Promises, Promises (Score:3)
Oh hey! This looks fascinating! (Score:1)
Now... (Score:1)
Re: (Score:2)
My current ISP still thinks I'm using MFM drives.
Re: (Score:1)
That's because of the type of pictures you download?
Bert
Re: (Score:2)
Not, it's because it's a 5 megabits per second [wikipedia.org] cable modem.
You can drop your nerd card in the little box on the way out.
Re: (Score:2)
I love ASCII art pr0n!
Re: (Score:2)
No Encryption required! (Score:1)
Good news everyone! (Score:4, Insightful)
If they can read it at least as fast as today's technologies, the power required to read/write data is roughly the same as today's drives and the manufacturing cost is also about the same, this is good news for everyone:
1. On the consumer side, cheaper drives per terabyte meaning cheaper home media servers
2. On the commercial side, a lot less energy required, i.e. no need for ultra-fast 15k RPM drives in servers, need up to 15 times fewer drives in server farms. This is BIG.
There is only one problem [xkcd.com].
Re: (Score:2)
"2. On the commercial side, a lot less energy required, i.e. no need for ultra-fast 15k RPM drives in servers, need up to 15 times fewer drives in server farms. This is BIG."
Probably not. Spindles == speed and redundancy. If you are looking at a data warehousing situation then maybe but if you are dealing with a lot of transactions you will still want as many spindles as you can afford.
Re: (Score:3)
We have 15k RPM drives because we need to move the sector to which we want to read/write at quicker to the actuator head. The slowest point isn't the transfer of data from head to platter, but a) moving the actuator arm and b) waiting for the correct sector to come around.
I'm not sure how much performance benefit using lasers could help since access time (moving the mechanical arm) is still on the order of ms.
Re: (Score:2)
Pay wall (Score:2)
If it's paywalled, it didn't happen!
Amphibious storage (Score:2)
As an added bonus the factory can continue to operate even if it's flooded [//to do: insert conspiracy theory here] as the lasers can then be attached to sharks.
This sounds like Perl (Score:2)
Re: (Score:2)
You mean like old 3.5" floppy disks!
Finally! (Score:2)
A classic tecnology updated!
http://www.national.com/rap/files/datasheet.pdf [national.com]
Congratulations! (Score:1)
Re: (Score:3)
Write without reading? (Score:3, Interesting)
Re: (Score:3)
No way to read it? (Score:1)
No man... (Score:1)
My head is spinning (Score:1)
60fs pulse, but what is the rep rate (Score:1)
Hmm (Score:1)
only detailed writing data with lasers; there's no word on how to read it.
Sounds like Windows' strategy: Crap the write to wherever on disk, and don't care about performance in reading it back. Why bother when read-time performance, when the user can defrag every day?
RDX devices (Score:2)
This is an old invention (Score:2)
How do they know it worked? (Score:2)
If they can't read it, how can they know that the lasers wrote successfully? Or does that mean they read it using conventional means?
Convar reads data from hard drives using lasers (Score:2)
Well obviously.... (Score:2)
"Unfortunately the York scientists only detailed writing data with lasers; there's no word on how to read it."
Use lasers. Duh. :)
-Charlie
Hot damn! (Score:2)
OK, my visual cortex is officially due for repair. I read the headline as "New Technique Promises Much Faster Hot Damn Write Speeds"
Re: (Score:2)
Lasers are currently used in reading optical media like CDs and DVDs... this is magnetic media
Re: (Score:2)
Re: (Score:2)
In the case of magnetic media, lasers write by literally burning sections of the disc and read by reading the reflections from burnt/unburnt sections to get bits (commercially pressed discs have bumps instead of burned sections, but are read in much the same manner). From what I recall, CD burners are separate lasers for burning and reading. Optical media just happens to use lasers for both reading and writing, but they're two different lasers (assuming I remember correctly).
I'm guessing here, but I'm assum
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
Indeed, if the disk could be also read the same way like MO disks, it would be promising: http://en.wikipedia.org/wiki/Magneto-optic_Kerr_effect [wikipedia.org]
Re: (Score:2)
Been done before. Chmod 222.